Herpes Simplex Virus sequencing shows stability in the genomes and variability in the technology

Read the full article on ScienceDirect

This research article is dedicated to the late Priscilla Shaffer

Text by Robin Colgrove


Though well-known to those slogging deep in the genome-sequencing trenches, less-appreciated aspects of NextGen sequencing merit wider exposure among general audiences. Technological advances have made “easy” sequences, well, easy -and rendered “difficult” sequences doable- but announcement that some genome is “sequenced” now often means mostly sequenced: the hard bits left for “later”. Also, advanced methods can now look back at sequences and isolates derived using “traditional” techniques to see if we really have what we thought. Finally, phenotypic data from strains characterized before current modalities can now be mined to better understand the links between genotype and phenotype.


Here, we employed a combination of meticulous, old-fashioned, pre-kit techniques for virus production, isolation and genome purification then NextGen sequencing and manual finishing of “difficult” herpes simplex virus genomes, whose abundant repeated sequences and high-GC content previously frustrated attempts at obtaining fully “polished” genomes. The KOS strain of HSV-1 and its derivatives have rich histories, both in the circumstances involved in their isolation and in the body of subsequent laboratory research on these standards. We describe these here and see what lessons can be drawn from a detailed look at the genomes.


KOS was isolated from a lip lesion by and from Kendall O. Smith at Baylor in 1964, selected for its attractive PFU/virion ratio. Phylogenetic analysis, though, suggests an Asian origin for the strain, perhaps from Dr. Smith’s military service in Korea.  It was established as a standard by the late Priscilla Shaffer, to whom our paper is dedicated and who supported our earlier efforts. KOS1.1 was picked over a decade and rounds of plaquing later by Hughes and Munyon in Buffalo, screening for growth at elevated temperature, demonstrating markedly reduced neurovirulence. Despite this, we see a striking lack of inter-strain variation: only five coding changes out of seventy-odd genes.  The few changes are intriguing, like the R->H mutation in the major glycoprotein, gB, in a region potentially influencing re-folding from pre- to post-fusion states. In contrast, comparison with Strain 17, the first reference HSV-1 genome sequence, shows numerous differences, such as short “frameshifts”, which in context were likely sequence errors from older technology.


Bringing together the history of standard viral strains with what is gleaned through modern techniques has interest in its own light and also helps understand both the promise and the limitations of our methods.


Table 4

Image legend

In contrast to the strikingly few coding changes between KOS and KOS1.1, comparison with the original HSV-1 reference sequence, Strain 17, published in 1988, shows multiple biologically unlikely short regions of amino acid sequence divergence due to paired, compensating “pseudo-frameshifts” at the nucleotide level,  likely resulting from base-calling off the old autoradiograms.

Introducing the author

Robin Colgrove

First author Robin Colgrove

About the research

History and genomic sequence analysis of the herpes simplex virus 1 KOS and KOS1.1 sub-strains

Robert C. Colgrove, Xueqiao Liu, Anthony Griffiths, Priya Raja, Neal A. Deluca, Ruchi M. Newman, Donald M. Coen, David M. Knipe

Virology, Volume 487, January 2016, Pages 215–221